Journal de bactériologie et parasitologie

Abstrait

Establishment of Valid Laboratory Case Definition for Human Leptospirosis

Marga G.A. Goris, Mariska M.G. Leeflang, Kimberly R. Boer, Marco Goeijenbier, Eric C.M. van Gorp, Jiri F.P. Wagenaar and Rudy A. Hartskeerl

Laboratory case definition of leptospirosis is scarcely defined by a solid evaluation that determines cut-off values in the tests that are applied. This study describes the process of determining optimal cut-off titers of laboratory tests for leptospirosis for a valid case definition of leptospirosis. In this case the tests are the microscopic agglutination test (MAT) and an in-house IgM enzyme-linked immunosorbent assay (ELISA) both on single serum and paired samples using a positive culture as the reference test in the Dutch population. The specificity was assessed using panels of sera from healthy donors, cases with known other diseases and non-leptospirosis cases with symptoms compatible with leptospirosis. Cases were divided into three periods corroborating the acute phase (1-10 days post onset of illness (DPO)), the early convalescent (11-20 DPO) and the late convalescent phase (>20 DPO). Cut-off titers for MAT and IgM ELISA were determined as 1:160 and 1:80 respectively for all three periods. These cut-off titers combined 100% specificity with a sensitivity that changed according to the stage of disease for both tests. The low sensitivities in the early acute phase are consistent with the dynamics of the humoral immune response. IgM ELISA yielded higher sensitivities compared to MAT in the acute and early convalescent stages. Moreover, the optimal sensitivity of MAT, the gold standard was < 82%, implying that a significant part of global cases is missed by this recommended test. MAT and IgM ELISA manifested partly complementary, resulting in a higher sensitivity when combining the results of these two tests. The availability of paired samples and of adequate clinical and epidemiological data are other parameters that will significantly increase the sensitivity of laboratory confirmation. This study enables fine-tuning of the current laboratory definition towards an improved case finding and implies that solid validation of laboratory parameters for case definition will improve both the diagnosis for individual patient care and for estimating the disease burden at a worldwide scale.